DESCRIPTION: (Applicant's Description) Colon cancer results from the accumulation of mutations in multiple genes that play a role in signal transduction, cell cycle control, and DNA repair. Relatively early in the process of colon tumorigenesis, cyclooxygenase-2 (COX-2), a gene coding for an arachidonic acid oxygenase, is induced and remains elevated through the transition from a benign to a malignant tumor. COX-2 makes bioactive arachidonate metabolites that function in a diverse array of intracellular and intercellular signaling pathways that control proliferation and apoptosis inter alia. The two COX isoforms are the molecular targets for nonsteroidal anti-inflammatory drugs (NSAIDs), which may account for the apparent chemopreventive activity of NSAIDs in human epidemiological and experimental animal studies. One of the products of COX-dependent arachidonic acid oxygenation is malondialdehyde (MDA). MDA is mutagenic and carcinogenic and reacts with DNA to form a series of covalent adducts. The principal MDA-DNA adduct is a pyrimidopurinone formed by reaction with deoxyguanosine, which we have termed M1G. This adduct is an efficient premutagenic lesion that induces transversions to T, transitions to A, and frameshift mutations. We hypothesize that COX-2 expression in colonic epithelial cells results in DNA damages a result of the production of MDA. This would provide a mechanism linking COX-2 expression with the generation of DNA damages leading to genetic mutations that may contribute to the progression of benign-to-malignant tumor cells. We propose to test this hypothesis by performing a series of studies of MDA and M1G generation by colonic cell lines and tumors. Specifically, we will: 1) determine the ability of a series of colon cell lines that express low or high levels of COX-2 to produce MDA and its fragmentation partner, 12-hydroxyheptadeca-5,8,10-trienoic acid (HHT); 2) determine the ability of a series of colon cell lines that express low or high levels of COX-2 to generate the MDA-DNA adduct M G; 3) determine the effects of a series of1 structurally diverse COX-1 and COX-2 inhibitors on eicosanoid biosynthesis, MDA generation, and M1G levels in cell lines that overexpress COX-2; 4) quantitate the levels of M1G in DNA of xenografts of high and low COX-2 expressing human tumors grown in nude mice and in normal colonic tissue from COX1-1 and COX-2 knockout mice; 5) quantitate the high levels of M1G in DNA of normal tissue or tumors from patients diagnosed with polyps or colon adenocarcinomas. These experiments will establish whether there is a link between COX-2 expression in colonic epithelial cells, an increase in DNA damage, and the prevention of DNA damage by NSAIDS.